1

2

David Schramm Symposium:

NEW VIEWS OF THE UNIVERSE

Recent Studies of Ultra High Energy Cosmic Rays

Alan WatsonUniversity of Leeds, UK

(regular KICP Visitor)

a.a.watson@leeds.ac.uk

3

Dave Schramm: 15 November 1997

(re- UHECR)

4

Outline:• Present Status of Detectors - Auger

• The Issues: i Changes to Hadronic Interaction Models

- inferences for mass composition

ii Energy Spectrum – is there a GZK-effect?

iii Arrival Directions - Clusters? BL Lac associations?

• Summary

5

Exposure and Event Numbers from various Instruments

km2 sr year Approximate rate > 10 EeV (km2 sr year)-1

AΩt N rate

AGASA: closed in January 2004: 1600 827 0.52 Scintillator arrayHiRes I: monocular ~5000 403 0.08 Fluorescence Detector(HiRes II: monocular

HiRes: stereo (PRELIMINARY) ~2500 ~500 0.20

Yakutsk: ~900 171 0.19 Scintillator plus air-Cherenkov light

Auger: data taking since Jan 2004 1750 444 0.25 Fluorescence plus water-Cherenkov

6

Water-Cherenkov detectors respond to muons, e±, γ

Fluorescence in UV →

11

Hybrid Approach of Auger Observatory

AND

300 – 400 nm

Nitrogen fluorescence

7

1056 surface detector stations deployed with 919 taking data (2 Dec 2005)

Three fluorescence buildings complete each with 6 telescopes

First tri-oculars in August

Status

CLF

8

θ~ 48º, ~ 70 EeV

Flash ADC tracesFlash ADC traces

Lateral density distribution

Typical flash ADC trace

Detector signal (VEM) vs time (ns)

PMT 1

PMT 2

PMT 3

-0.5 0 0.5 1.0 1.5 2.0 2.5 3.0 µs

18 detectors triggered

9

Lateral density distribution

θ~ 60º, ~ 86 EeV

Flash ADC traces

Flash ADC Trace for detector late in the shower

PMT 1

PMT 2

PMT 3

-0.5 0 0.5 1.0 1.5 2.0 2.5 3.0 µs

35 detectors triggered

Much sharper signalsthan in more vertical events leads toν- signature

10

Pixel geometryshower-detector plane

Signal and timingDirection & energy

FD reconstruction

11~4 x 1019eV

Stereo-Hybrid Event

x

x x

12

The Central Laser Facility of the Pierre Auger Observatory

355 nm, frequency tripled, YAG laser, giving < 7 mJ per pulse: GZK energy

13

ti

Geometrical Reconstruction

14

Angular and Spatial Resolution from Central Laser Facility

Laser position – Hybrid and FD only (m)Angle in laser beam /FD detector plane

Mono/hybrid rms 1.0°/0.18° Mono/hybrid rms 566 m/57 m

15

A Big Event - One that got away!

Shower/detector plane

Fluorescence Mirror

Energy Estimate

>140 EeV

19 April 2004

16

(ii) Muon Content of Showers:-

N (>1 GeV) = AB(E/A)p (depends on mass/nucleon and model)

N(>1 GeV) = 2.8A(E/A)0.86 ~ A0.14

So, more muons in Fe showers

(i) Variation of Depth of Maximum with Energy

Methods of Inferring the Primary Mass

HADRONIC M

ODELS REQUIR

ED

FOR INTERPRETATIO

N –

New m

odel, Q

GSJET II,

dis

cuss

ed

at IC

RC

******

******

******

******

Xmax

Log E

Limiting bound = 2.3 X0 g cm-2 per

decade (Linsley 1977)

p

Fe

17

Regions of most interest for shower modelling

Pseudo-rapidity

Multiplicity

Energy distribution

18

New hadronic model: QGSJETII Heck and Ostapchenko ICR 2005

Multiplicity vs. Energy

19Heck and Ostapchenko: ICRC 2005

Xmax vs. Energy for different models compared with data

*

20

Hooper, Sarkar and Taylor 2005

Assumption: Fe with E-2 with sharp cut-off at 1022 eV

21

Spectrum measurements: Issues of concern

1: SURFACE DETECTOR ARRAYS (e.g. AGASA, Yakutsk)

APERTURE:

- relatively easy to determine

ESTIMATION OF PRIMARY ENERGY:

- mass assumption required

- hadronic interaction model must be assumedfor which systematic uncertainty in UNKNOWABLE

- QGSJETII model will lead to revisions

22

2. FLUORESCENCE DETECTOR (e.g. HiRes):

ENERGY ESTIMATES depend only weakly on assumptions about models and mass

BUT determination of energy requires- atmospheric corrections for each event- Cherenkov light subtraction (< 25% used)

APERTURE is difficult to measure- does not saturate- depends on atmosphere- mass of primary- models- spectral shape

so, aperture can be systematically uncertain

23

3. Hybrid Detectors (e.g. Auger)

ENERGY CALIBRATION of size parameter measured by surface detectors is made with fluorescence detectors on carefully selected sample of events:

- long tracks in atmosphere: > 350 g cm-2

- Cherenkov light contamination: < 10% (Auger criteria)

HIGH STATISTICS from surface array

APERTURE: well-defined

24Pierog et al. ICRC 2005

Ratio of total energy to electromagnetic energy for fluorescence detector

Etot/Ecal

log 10 Etot (eV)

1.10

~7%

25

Auger Aperture

AGASA aperture

Rel

ativ

e A

pert

ure

1.0

0.5

Energy (eV)1018 1019

26

HiRes Monocular Spectra: ICRC

The HiRes grouphave yet to releasea stereo spectrum.

It will have hour-by-houratmospheric correctionsusing monitoring data

Should also help to resolvethe aperture uncertainties- at least at smalldistances

Choice of data usedin the fit is entirelysubjective and nopropagation of Eerrors into y-direction

27

Auger Energy Determination: Step 1

The detector signal at 1000 m from the shower core

– called the ground parameter or S(1000)

- is determined for each surface detector event using the lateral density function.

S(1000) is proportional to the

primary energy.

The energy scale is determined from the data and does not depend on a knowledge of interaction models or of the primary composition – except at level of few %.

Zenith angle ~ 48º

Energy ~ 70EeV

28

Auger Energy Determination: step 2

log S(1000) from SD

log (

E/E

eV

) fr

om

FD

10EeV

1 EeV

Hybrid Events with STRICT event selection:

aerosol content measured

track length > 350 g cm-2

Cherenkov contamination <10%

29

Spectrum measured with Auger Observatory

The function is

F=(30.9±1.7)(E/EeV)-1.84+/- 0.03

with χ 2 = 2.4 per degree of freedom

Issues of aperture, massand hadronic interactionsunder control – systematic uncertainties being assessed

S(1000)Fluorescence YieldAbsolute FD calibrationS(1000) to energy – and limited statistics

systematicuncertainty

30

))ln(

(100

FEd

dI

F

Percentage Deviation from the Power-Law Fit

31

Summary Spectrum above 2 EeV

aaw/Sept 2005

32

Deviations of data from E-3 line through first point of Auger data

aaw/Oct 2005

33

HiRes stereo events > 10 EeV plus AGASA events above 40 EeV

HiRes Collaboration: ICRC 2005: Westerhoff et al.

Candidate Clusterα = 169.0 δ = 56.2

Analysis uses likelihood ratio method:

p = 43% for 271 HiRes and 47 AGASA events

NB: Previously, Finley and Westerhoff had shown previouslythat AGASA clustering was statistically unconvincing

- Clustering is very far from being established

34

HiRes does see

correlations with BL Lacs:Veron 11th Catalogue:

178 objects with magnitude < 18

Claim: excess number of BL Lacs seen near HiRes events > 1010 GeV, consistent with the HiRes angular

resolution ~ 0.6º

GOOD ANGULAR RESOLUTION

see 11 pairs < 0.8º and expect ~ 3,

⇒ probability ~ 5×10-4

But these BL Lacs are hundreds of Mpc distant!

Few % of primaries must be neutral @ 1010 GeV!!

Gorbunov et al (2004)

Westerhoff et al ( 2005)

35

2 x 10-4

5 x 10-4

10-3

2 x 10-4

10-5

2 x 10-4

Finley and Westerhoff ICRC 2005

Group is awaiting independent data set recorded post January 2004up to closure in March 2006 before making any claims. They have concerns about ‘over tuning’

Summary of BL Lac Searches

36

Auger sees no concentration of events along the Galactic or Super-Galactic planes

(Antoine Letessier-Selvon, ICRC 2005)

Auger Observations show NO concentration of events along

Galactic or Super-Galactic Plane

A: 1 – 5 EeV: Galactic Plane

B: > 5 EeV: SGP

C: > 10 EeV: SGP

37

If highest energy particles were protons, and there is no anisotropy, exotic origin ideas have to be invoked

• Decay of super-heavy relics from early Universe (or top down mechanisms) Wimpzillas/Cryptons/Vortons

• New properties of old particles or new particles

• Breakdown of Lorentz Invariance

Predictions: dominance of neutrinos and photons

38Picture by Dieter Heck

On-set of LPM effect

γ + B → e++ e-

39(Risse for Auger Collaboration, ICRC 2005)

… the highest energy particles seem NOT to be dominantly photons

Constrains (but does not yet rule out) ‘top-down’ models of UHECR origin

Photon limit from Auger observations

40

Ideas to explain the Enigma• Decay of super heavy relics from early Universe (or top down mechanisms)

Wimpzillas/Cryptons/Vortons

Few photons: <26% at 1019 eV (Auger claim)

Model predictions have changed

Is there need for exotic explanations?

or is it ‘simple’?• Are the UHE cosmic rays iron nuclei at source?

• Are magnetic field strengths really well known?

41

Summary: I

• Arrival Directions:

No convincing evidence for anisotropyPossibility of BL Lac associations could

be clarified in ~ 2 years

• New Hadronic Interaction Model:

suggests that there could be a heavier mass > 10 EeV than has been supposed by many in the past

Heavier mass would ease acceleration, isotropy and spectrum issues

BUT – Nature may have surprises to show at the LHC

42

Energy Spectrum: Auger: ~ 5 to 7 X AGASA by 2007Spectrum that is largely mass and model independent

AGASA/HiRes/Auger differences could – possibly – be understood through combination of improved understanding of HiRes aperture (composition/spectrum/ hadronic model and stereo data)AND different models and mass assumptions by AGASA

ALL GROUPS HAVE REPORTED EVENTS ABOVE 100 EeV

QUESTION: WHAT IS THE DETAILED SHAPE OF THE SPECTRUM?

Summary: II

43

Thanks to all of my Auger colleagues

Spokesperson:Alan Watson

Czech Republic ArgentinaFrance Australia Germany BrasilItaly Bolivia*

Netherlands Mexico Poland USASlovenia Vietnam* Spain United Kingdom

*Associate Countries

~250 PhD scientists from 63 Institutions and 15 countries

44

45

Electromagnetic Acceleration

• Synchrotron Acceleration Emax = ZeBRc

• Single Shot Acceleration Emax = ZeBRc

• Diffusive Shock Acceleration Emax = kZeBRc, with k<1

Shocks in AGNs, near Black Holes……

46

Hillas 1984 ARA&A B vs R

Magnetars?

GRBs?

47

Some properties of 20 highest energy events :

Curvature

Thickness of shower discFall-off of signal size withdistance from axis

Showers of 30 EeV are just likeshowers at 1 EeV – but bigger

48

Resolution of Core Position

Hybrid – SD only core position

Hybrid DataLaser Data

Core position resolution:Hybrid: < 60 m Surface array: < 200 m

Laser position – Hybrid and FD only (m)

-500

+500

501

rms spread ~ 570 m for monocular fit

49

Angular Resolution

Surface array Angular resolution (68% CL)<2.2º for 3 station events (E< 3EeV, θ < 60º )< 1.7º for 4 station events (3<E<10 EeV)< 1.4º for 5 or more station events (E>10 EeV)

Hybrid Angular resolution (68% CL) 0.6 degrees (mean)

Hybrid-SD only space angle difference

Hybrid Data

Angle in laser beam /FD detector plane

Laser Beam

Entries 269

σ(ψ) ~ 1.24º

Resolution using a centrally positioned laser

50

Ratio of Aperturescomputed with SIBYLLand QGSJET

Sensitivity of HiRes II aperture to shower model

Zech et al. HiRes Collaboration: ICRC 2005

More statistics needed andup-dated model needs to be used.

Mass assumption has onlybeen explored at ± 5% of anassumed proton fraction

51

Arrival Direction Studies

52

53

54

• Fit to power law.

• Single index gives poor χ2

• Evidence for changing index

1019 1020

HiRes Stereo Flux

Springer et al. ICRC 2005F

lux x

1029

log E

55

56

Association with BL Lacs?

Initial claims by Tinyakov et al. – but disputed by Evans et al and others

217 HiRes Stereo events above 10 EeV

σ = 0.4 deg, so that 68% of events would lie within θ = 1.52 σ

This is an impressive angular accuracy

Tinyakov et al. conclusion for m<18 confirmed – but same data set of events and same 157 BL Lacs

BUT for E> 40 EeV, HiRes shows a deficit in the correlation

Presumably primaries are neutral because of anticipatedmagnetic field deflections – worth looking at lower energies.

57

105573

12 13 14 km

10 9 8 km

Hybrid events are equivalent to stereo- events and superior to monocular events

Observations with real

showers confirm the

results from Central

Laser Facility

58Heck and Ostapchenko: ICRC 2005

SIBYLL

Muon Number Ratio for different models and masses

10% reduction in predicted muon numberleads to ~ x 2 increase in the average mass –depending on model details

10%

59

Original Claim (2003):

“Consistent with proton dominant component” –

must be revised

19 19.5 20 20.5

Log(Energy [eV])

−2

−1

0

1

Log(

Muo

n de

nsity

@10

00m

[m–2

])Muon measurements with the AGASA array

60~4 x 1019eV

Stereo-Hybrid Event

x

x x

61

Surface Array 1600 detector stations 1.5 km spacing 3000 km2

Fluorescence Detectors 4 Telescope enclosures 6 Telescopes per

enclosure 24 Telescopes total

CLF

62

HiRes I and HiRes II

63

ti

Geometrical Reconstruction